A central problem of democracy is the aggregation of divergent individual inputs into overall collective decisions. Social-choice-theoretic impossibility results famously demonstrate the intractability of a large class of such aggregation problems. This paper develops a taxonomy of two concepts of agreement, agreement at a substantive level and agreement at a meta-level, and discusses the escape-routes these concepts open up from the impossibility problems of social choice. Specifically, two contexts of democratic aggregation are addressed: first, the familiar context of preferences, and second, the largely unexplored context of sets of judgments over multiple interconnected propositions. Drawing on some recent developments in social choice theory and democratic theory, I will defend the view that, when agreement is conceptualized in democratic theory and when it is sought in democratic practice, more emphasis should be placed on agreement at a meta-level than is commonly done. Finally, I will address the more general question to what extent it is acceptable for the stability of institutions for democratic aggregation to be dependent on specific empirical contingencies.